Dear Player, Day 2️⃣ & the hype is still peaking! 🚀 Don't let the best vibe pass you by. Swing by Booth SH-2714 before doors close! Yostar Booth SH-2714 | July 2-5, 2026 | Los Angeles Convention Center, CA #Yostar# #StellaSora# #AzurLane# #Arknights# #MahjongSoul# #AnimeExpo# #AX2026#

UBS initiated $CRWV at Buy with a $120 price target citing “outstanding AI compute demand signals” and saying revenue per GW will march higher. The firm also says leverage and credit concerns may be peaking while the stock still trades at just ~3x 2027 revenue.

Glassnode: Bitcoin Remains in Deep Value, but Bottom Confirmation Signals Are Still Missing According to Glassnode, Bitcoin has traded below the True Market Mean of $76,600 and the short-term holder cost basis of $72,200 for about five months, keeping it in deep value territory while a confirmed bottom has yet to form. On-chain data shows long-term holder loss realization has risen from 15% of total realized value in early February to 43%, with daily realized losses peaking at $280 million, the highest level since December 2022. ETF outflows have narrowed from $193 million to $88.9 million per day but remain negative, while daily ETF trading volume of $650 million to $950 million is about 80% below the October 2025 peak, suggesting institutional demand has not stabilized. In derivatives, the options open-interest put/call ratio has fallen to 0.56, its 2026 low, indicating reduced short demand, though options skew still reflects demand for downside protection. Glassnode said Bitcoin may be in the later stage of a bear-market bottoming process, but confirmation would require cooling long-term holder selling pressure, stabilization in ETF flows, and a recovery above key cost-basis levels.

$MU just posted a blowout quarter. Tens of billions added to the top line, massive margin gains across every segment, and growth accelerating both year-over-year and quarter-over-quarter. Any concerns about "peaking margins" are overblown, @BrianSozzi says, especially with the stock still trading at just 8 times forward earnings. 📈 😮

New Yorkers, here’s what you need to know about this week’s nor’easter: The storm is expected to pass offshore, keeping the heaviest rain east of New York City. But we should still prepare for gusty winds, rain, and some high-tide flooding beginning today, peaking Friday, and potentially continuing through Sunday. If you live near the coast, move your car to higher ground or away from streets that regularly flood. Secure loose outdoor items, including furniture and garbage containers, that could be picked up by strong winds. Storm tracks can change, so please stay informed and be prepared. For emergency alerts, text NOTIFYNYC to 692-692.

As Google pushes AI deeper into Search, more users are seeking simpler, more traditional alternatives. DuckDuckGo reported that visits to its AI-free search page jumped 22.7% on average last week, peaking at 27.7%. The company also saw an 18% increase in US mobile app installs. This suggests not everyone wants AI-generated answers and many users still prefer clean results and more control over browsing. Though Google still dominates with roughly 85% US market share compared to DuckDuckGo’s 2%

Raymond James Raises $NVDA PT to $352 from $330 - Strong Buy Analyst comments: "NVIDIA has increased its visibility in CPUs even if they are not a new part of the story. However, CPUs represent a low-single-digit percent of NVIDIA’s sales — roughly 3% today. In the past, NVIDIA’s CPUs were tightly tied to its GPUs, but that should change, particularly with agentic use cases. While NVIDIA’s CPU revenue is becoming material within the CPU market, we expect it to reach approximately 5% of total revenue by CY28 in our model — still a single-digit portion, yet the growth is the fastest among the elements we model. We have extended our model through FY29/CY28 and refined our segment assumptions. Among the surprising conclusions from our market analysis is the prospect that NVIDIA could become the world leader in CPU revenue within several years. We have been disappointed with the stock’s year-to-date performance, with the shares up only 12%, largely matching the S&P 500 Index. The stock trades at a CY27 GAAP P/E of less than 15x, which is below the S&P 500’s 18.6x. The discount strikes us as fundamentally illogical considering that sales and net income growth still exceed 20% in our CY28 estimates. Typically, a company with sustainable growth, barriers to entry (e.g., CUDA, leading GPU performance), and healthy free cash flow trades at multiples above the overall market. Factors that may weigh on NVIDIA’s shares include a $5.2 trillion market cap that may limit further exposure, fear that the AI boom turns to a bust, decelerating growth, and peaking growth and margins. Our $352 price target, up from $330, is based on a 22x multiple applied to our new CY28 estimates. This represents a premium to the S&P 500’s ~19x CY27 multiple. Considering NVIDIA’s industry leadership and strong growth, the market premium is conservative. The median NTM P/E has been 35x over the past five years." Analyst: Simon Leopold

Merkle3s Weekly Recap: US-Iran ceasefire framework lands as oil peaks; Warsh's FOMC debut all but certain to hold — World Cup ignites prediction markets, Hyperliquid SPCX perps hit $1.4B on day one 1️⃣ Macro: Rates Peak, Volatility Persists ➡️ The US and Iran reached a framework to end hostilities, with formal signing set for Friday in Switzerland and the Strait of Hormuz expected to reopen shortly after. Shipments through the strait have already rebounded and supply-demand stays tightly balanced — even if the deal hits a snag, oil likely holds below $90, with diminishing market impact ➡️ May headline and core CPI both came in below expectations MoM. With US gasoline prices falling, May is likely the year's inflation peak, and June CPI looks set to turn negative MoM. Two hikes this year are unlikely — a single cut remains on the table. The case for further upside in long-end Treasuries has evaporated, and the center of gravity will drift lower. This sets up a year-defining allocation window for gold ➡️ The June 18 FOMC is Chair Warsh's first, with markets pricing a 98%–99% chance of holding at 3.50%–3.75%. Watch for dot-plot upward revisions that could rattle sentiment, but any bounce is likely still a topping pattern ➡️ US equities have entered a high-volatility regime. South Korea's KOSPI triggered a circuit breaker last week (opening down over 8%), driven by excessive semiconductor leverage and retail-concentrated positioning. The technical correction looks complete for now, but the underlying issues remain unresolved — a final 10%–15% drawdown can't be ruled out, though this stops short of a bear market call 2️⃣ Prediction Markets: World Cup Kicks Off a Volume Surge ➡️ The 2026 FIFA World Cup (co-hosted by the US, Canada, and Mexico — 48 teams, 104 matches) is underway. Polymarket's World Cup winner market alone has topped $1.8B, with 878 related markets collectively breaking $2B — a platform record for sports prediction markets ➡️ Kalshi's equivalent winner contract stands at roughly $121M, about 1/15th of Polymarket's. Still, Kalshi's ~$6B 30-day volume leads Polymarket's $4B; on OI, Kalshi sits at ~$833M vs. Polymarket's ~$486M 3️⃣ Perp DEX: Range-Bound Volume Pulls Back, HIP-3 Share Keeps Climbing ➡️ Crypto traded in a tight range last week, with BTC oscillating between $61,000–$63,500 and overall volume declining. BTC now sits at roughly $65,700, with total crypto market cap at ~$2.25 trillion ➡️ Hyperliquid's HIP-3 (TradFi asset perps) volume keeps rising, reaching as much as 64% of total platform volume within a single 24-hour window. Cumulative equity perp volume has topped $10B. The S&P 500 perp alone has reached $721M in OI, generating over $918K in fees from a single market 4️⃣ Hyperliquid × SpaceX IPO: Price Discovery Narrative Validated Again ➡️ After SpaceX listed last Friday, the $SPCX perpetual on Hyperliquid saw $1.4B in day-one volume, 30% of the HIP-3 total. Pre-listing volume exceeded $450M across 917,000 trades, with OI peaking above $100M ➡️ Hyperliquid's pre-IPO price discovery on SpaceX proved accurate, closely tracking the actual opening price and further cementing the narrative. Unlike Robinhood, Coinbase, and Binance, which saw outages or refunds during the IPO window, Hyperliquid executed without disruption, reinforcing its 24/7 uptime and high-concurrency edge 5️⃣ RWA: The SPCX Tokenized Issuance Controversy ➡️ xStocks issued tokenized SpaceX shares (SPCXx) via Kraken, Bybit, and other platforms, supporting USDC subscriptions with 1:1 real equity backing and no lockup — allocated shares become immediately tradable onchain. This marks an early experiment in tokenized primary-market IPO issuance ➡️ Issues in xStocks' issuance and allocation process hit the vast majority of participating exchanges, leaving large numbers of users unable to subscribe as expected — and community frustration keeps building. The SpaceX IPO was heavily oversubscribed, and retail allocations through traditional channels are expected to be minimal ➡️ The core distinction between tokenized shares and perpetuals has sparked debate: the former carries 1:1 backing with real equity rights, the latter a pure price bet with no voting power. Allocation fairness, instant liquidity vs. stability, and whether retail end up as exit liquidity have become the main points of contention.

Who, in their wildest delusion, is trying to weaponize 'overcapacity' — while the 'China Dividend' is actually powering the world? 谁在妄言“产能过剩”？“中国红利”正在驱动全球！ Recently, the phrase "China's overcapacity" has been appearing with striking frequency. In North America and Europe, it comes up in think tank reports, trade hearing testimonies, and political speeches — all in remarkably uniform tone. But interestingly, the louder the outcry, the more the actual market moves in the opposite direction. This summer, European consumers rushed to buy Chinese air conditioners — in the first half of this year, China's air conditioner exports to the EU surged 43.2% year-on-year, hitting a record high. Meanwhile, developers worldwide are flocking to Chinese large-language models — on OpenRouter, a multi-model aggregation platform, all top-five models by weekly calls come from Chinese companies. One vote is cast with wallets; the other, with code. If there really were "overcapacity," why would the market still be scrambling for these products? The so-called "China overcapacity" is, in the final analysis, a false proposition that cannot withstand market scrutiny. Recently, China's Ministry of Commerce released a 12,000-word position paper titled "The So-Called 'Overcapacity' Issue: China's Position," which gets to the root of the "overcapacity" narrative. The document systematically responds to relevant international discussions and firmly pushes back against the erroneous argument that "Chinese subsidies cause overcapacity." One sentence in the paper cuts straight to the heart: "A large export volume and a trade surplus of a country are the inevitable outcome of the international division of labor in the era of economic globalization." The weight of this statement becomes particularly clear when viewed against the broader coordinates of global economic development. In psychology, there is a concept called "attribution theory": people tend to make "internal attributions" for their successes — crediting their own efforts — but lean toward "external attributions" when facing problems, shifting blame to outside factors. This tendency is especially pronounced among trade protectionists. When their industries fail to keep pace with the times, fall behind, or suffer from domestic hollowing-out, they point fingers outward and blame China. When 80% of U.S. chips are exported and two-thirds of Boeing aircraft are sold overseas, that is touted as industrial strength. When the EU runs hundreds of billions of dollars in trade surpluses year after year in automobiles, pharmaceuticals, and cosmetics, that is called normal international trade. Yet when China exports a lot, why is that suddenly "overcapacity"? Why is that "dumping at low prices"? Different yardsticks yield different conclusions — this is double standards in action. What is more telling is that by now, the "overcapacity" hype has long ceased to be a purely economic issue; it has been fully reduced to a tool of geopolitical rivalry in the West. The U.S. enacted the Inflation Reduction Act, doubling down on exclusionary local protectionism; the EU introduced the Industrial Accelerator Act, directly tying subsidies to local production. They pay lip service to fair competition, but in reality, they are erecting targeted trade barriers — a classic case of saying one thing and doing another. No matter whether the trade protectionists in North America and Europe are genuinely misguided or willfully obtuse, China's attitude remains clear and steadfast: we openly embrace healthy global market competition, but we will never allow normal commercial rivalry to be deliberately escalated and twisted into geopolitical confrontation. If we cut through the fog of rhetoric and take an honest look at the reality of international trade, we will see that the sustained growth of China's exports is rooted in the genuine global demand for green transition and industrial upgrading. Take the new energy sector as an example: Chinese photovoltaic modules now hold over 90% of the global market share, and Chinese power batteries account for more than 70% of global shipments. This is not ineffective excess capacity piling up in the domestic market; rather, it is the large-scale industrial advantages of China precisely matching the urgent global need for green transformation. More importantly, the global demand for new energy is far from peaking. The International Energy Agency projects that from 2025 to 2030, global installed renewable energy capacity will double, with nearly 80% of that growth coming from solar photovoltaic. By 2030, annual global sales of new energy vehicles are expected to reach ten times the 2023 level. Therefore, China's industrial expansion overseas has never been a "market shock" — it is a development dividend that benefits the entire world. The numbers speak for themselves: over the past decade and more, China's contribution to global economic growth has remained steady at around 30%. China has been the world's second-largest importer for 17 consecutive years and is a major export destination for nearly 80 countries. It has implemented zero-tariff treatment for 63 countries and is the world's first major economy to offer zero-tariff coverage for all African countries that have diplomatic relations with it, as well as for all least-developed countries that have diplomatic ties with it. China is the only country in the world that hosts an International Import Expo, which has been successfully held for eight sessions, with cumulative intended transaction volume exceeding $580 billion. During the 14th Five-Year Plan period, China's cumulative imports surpassed $90 trillion. The trade surplus may stay in China, but the industrial dividends, jobs, and development opportunities have genuinely benefited the entire world. Trade barriers may block Chinese products, but they cannot stop the advance of technological progress, the momentum of industrial rise, or the market's choice as expressed through its own votes: China is not only an increasingly robust "world factory" but also a vibrant "world market." China's modern industrial development is not "China Shock 2.0" for the world — it is "China Opportunity 2.0." 最近，"中国产能过剩"这六个字，出现的频率有点高。在北美与欧洲，智库提，贸易听证会提，政客演讲也提，调子出奇地一致。但有意思的是，嘴上喊得越凶，真实的市场，越是朝着相反的方向走。 今年夏天，欧洲的消费者争相抢购中国空调 —— 上半年中国对欧盟空调出口额同比暴涨 43.2%，创下历史新高。全球的开发者追捧中国的大模型 —— 多模型聚合平台 OpenRouter 的周度调用量榜单上，排名前五的产品全部来自中国企业。一个用钱包投票，一个用代码投票。 如果真的"过剩"，市场为什么还在抢？所谓"中国产能过剩"，说到底，是个经不住市场检验的伪命题。 最近，中国商务部发布了一份1.2万字的立场文件——《关于所谓"产能过剩"问题的中方立场》，一口气把“产能过剩论”的根给刨了。这份文件系统回应了国际上的相关讨论，正面回击了"中国补贴造成产能过剩"的错误论调。 文件里有一句话，直击要害："一个国家出口规模大、存在贸易顺差，是经济全球化下国际产业分工造就的必然结果。"这句话的分量，放在全球经济发展的坐标系里看，会格外清楚。心理学有个“归因理论”：人们看待成绩时喜欢“内部归因”，认为是自己努力的结果；面对问题时则倾向“外部归因”，把矛盾归咎于外部因素。在贸易保护主义者身上，这种现象尤其明显。一些国家的产业发展跟不上时代了，落伍被淘汰了，或者是国内产业空心化了，就把锅往外甩，说是中国的问题。 美国芯片80%用于出口，波音飞机三分之二销往海外被称作产业优势；欧盟汽车、医药、化妆品常年保持数百亿贸易顺差被叫做正常国际贸易。同样是出口多，到了中国这里，怎么就成了"产能过剩"？怎么就成了"低价倾销"？标准不同，结论当然不同，这就是双重标准。 更值得深思的是，炒到现在，"产能过剩"早就不单纯是一个经济问题了，它已经彻底沦为西方地缘博弈的工具。美国出台《通胀削减法案》，大搞本土排他性保护；欧盟推出《工业加速器法案》，把补贴和本地生产直接绑定。嘴上标榜的是公平竞争，实际构筑的却是针对性的贸易壁垒，典型的“说一套，做一套”。 不管北美与欧洲的贸易保护主义者是真糊涂还是装糊涂，中国的态度始终清晰而坚定：我们坦然接受全球良性市场竞争，但绝不允许正常的商业竞争被刻意升级、扭曲成地缘政治对抗。 拨开舆论的迷雾，正视国际贸易的现实就会发现，中国出口的持续增长，根源在于全球绿色转型和产业升级的真实刚需。以新能源产业为例，中国光伏组件全球市场占有率超过 90%，动力电池全球出货量突破 70%。这不是国内市场积压的无效过剩，而是中国产业的规模化优势，精准匹配了全球绿色转型的迫切需求。更关键的是，全球新能源市场的需求，远远没有触顶。国际能源署预测，2025年到2030年，全球可再生能源装机总量将翻一番，其中近80%的增量来自太阳能光伏；到2030年，全球新能源汽车的年销量，将达到2023年的十倍。所以，中国产业出海从来不是什么 "市场冲击"，而是普惠全球的发展红利。 数据就摆在那里：过去十多年，中国对世界经济增长的贡献率常年稳定在30%左右。中国进口规模连续17年居全球第二，是近80个国家的主要出口目的地；对63个国家实施零关税，是全球首个对所有非洲建交国和所有建交的最不发达国家实现全覆盖零关税的主要经济体；中国是全球唯一举办国际进口博览会的国家，已成功举办8届，累计意向成交额超5800亿美元；“十四五”时期累计进口规模超90万亿元…… 贸易顺差留在中国，但产业红利、就业岗位、发展机遇，实实在在惠及了全世界。贸易壁垒，可以挡住中国的产品，却挡不住技术进步的脚步，挡不住产业崛起的大势，更挡不住市场用脚投票的选择：中国不仅是日益强大的“世界工厂”，更是活力旺盛的“世界市场”。中国现代化产业发展对世界不是“中国冲击2.0”，而是“中国机遇2.0”。 #China# #Jiangxi# #世界经济看中国# #赣出新精彩#

The 4 BACKTESTING TECHNIQUES behind WINNING Strategies: i've spent the last 2 years running backtests on everything from mean reversion setups to volatility arbitrage to prediction market signals some strategies survived and tbh most of them died and the difference was never the strategy itself, it was how i tested it a backtest is not proof your strategy works, it's a stress test to see how easily it breaks here are the 4 techniques i've actually run, what worked, what broke AND what i still use --------------- technique 1: standard in-sample / out-of-sample split verdict: broken by default, NEVER TRUST THIS the setup is easy, take 5 years of data and train on the first 4, test on the last 1 the problem is subtle - every time you tweak the strategy and re-run, you're peeking at the test data and after 30 iterations your "out-of-sample" is FULLY contaminated the first strategy i ever backtested was a simple pairs trade between two energy stocks that showed a Sharpe of 2.1 on the standard split, so i deployed $2,000 of my own money and lost 40% of it in 3 months going back later i realized i'd re-run that backtest 47 times during tuning, the test data was never really untouched use this only for a quick first look, NEVER as the final validation --------------- technique 2: walk-forward validation verdict: the real workhorse, this is what i actually use instead of splitting once, you slide a window through the data train on 2018-2020, test on 2021 train on 2019-2021, test on 2022 keep sliding each test window is data the model has never seen and you get 5 or 6 test periods instead of JUST ONE what this catches: > strategies that only worked in one regime (the pattern shows up immediately) > parameters that shift wildly when retuned (unstable strategy, red flag) > strategies that survive across every window (this is real edge) at our fund we killed a stat arb strategy that showed Sharpe 2.4 on a standard split, but walk-forward revealed it worked beautifully in 2019-2020 and completely died in 2021-2022, the regime had shifted underneath us and it saved us months of losses but this is slower and more painful than a standard split and it's also the reason institutional backtests match live P&L :) --------------- technique 3: purged k-fold cross-validation verdict: fixes a hidden bug in walk-forward financial data has memory, today's price is not independent of yesterday's when your training window ends on december 31 and your test window starts january 1, information leaks across that boundary and your Sharpe looks better than it should purged k-fold fixes this, Marcos Lopez de Prado covers it in Advances in Financial Machine Learning the idea is simple: > split data into folds like standard cross-validation > when a fold is used for testing, remove the adjacent observations that overlap in time > this eliminates the leakage a QUANT friend of mine who runs an ML-based factor model showed me his numbers before and after adding purging, Sharpe dropped from 1.9 to 1.4 on the same strategy with the same data and the extra 0.5 was pure leakage he didn't know he had use this when you're training ML models on financial data, the leakage in tree-based models is brutal without it --------------- technique 4: monte carlo trade shuffling verdict: the reality check that saves capital EVERY SINGLE TIME your backtest shows one sequence of trades, Monte Carlo randomizes the order and runs it thousands of times why this matters: > your backtest might have gotten lucky with sequencing, what if the drawdown happened in month 2 instead of month 10 > the max drawdown you observed is one path, Monte Carlo shows the full range > the 5th percentile drawdown is often 2 to 3 times worse than what you saw few months ago (during the hype of 15-min BTC markets) i built a systematic prediction market strategy that showed 12% max drawdown across 18 months of backtest, but before deploying i ran Monte Carlo with 10,000 shuffled sequences and the 5th percentile scenario showed a 34% drawdown ofc i didn't deploy at full size, i sized it at 25% of what i originally planned and three months in the strategy hit a 22% drawdown, but the smaller size meant i could hold through it and the strategy recovered to finish the year up 31% Monte Carlo is why i stayed in that trade instead of blowing up --------------- what i actually use in production NOW: > walk-forward validation as the primary test > purged k-fold when the strategy uses ML models > Monte Carlo shuffling on strategies that survive both, before any real capital > standard in-sample/out-of-sample only for the very first pass (rarely tho) if your backtest is designed to make you feel good, it's designed to LOSE you money for real.